selective loss of sarcolemmal nitric oxide synthase in becker muscular dystrophy

Se lec t ive Loss o f S a r c o l e m m a l N i t r i c O x i d e Synthase in B e c k e r M u s c u l a r D y s t r o p h y

By Daniel S. Chao,* J. Rafael M. Gorospe,r E. Brenman,* Jill A. Rafael,~ Matthew E Peters,ll Stanley C. Froehner, ll Eric P. Hoffman,~ Jeffrey S. Charnberlain,~ and David S. Bredt*

From the *Department of Physiology and Program in Biomedical Sciences, University of California at San Francisco School of lVledicine, San Francisco, California 94143-0444; the *Department of Molecular Genetics and Biochemistry, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15261; the -~Department of Human Genetics, University of Michigan School of Medidne, Ann Arbor, Michigan 48109-0618; and the IIDepartment of Physiology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599

S u m m a r y

Becker muscular dystrophy is an X-linked disease due to mutations of the dystrophin gene. We now show that neuronal-type nitric oxide synthase (nNOS), an identified enzyme in the dystrophin complex, is uniquely absent from skeletal muscle plasma membrane in many human Becker patients and in mouse models of dystrophinopathy. An NH2-terminal domain of nNOS directly interacts with cxl-syntrophin but not with other proteins in the dystrophin complex analyzed. However, nNOS does not associate with cll-syntrophin on the sarcolemma in transgenic mdx mice expressing truncated dystrophin proteins. This suggests a ternary interaction of nNOS, ~l-syntrophin, and the central domain of dystrophin in vivo, a conclusion supported by developmental studies in muscle. These data indicate that proper assembly of the dystrophin complex is dependent upon the structure of the central rodlike domain and have implications for the design ofdystrophin-containing vectors for gene therapy.

M utations of the X-linked dystrophin gene are the most common cause of inherited muscular dystro-

phy and affect ~1:3,500 male births (1). Duchenne muscular dystrophy (DMD) I, the more common and more severe form of the disease, is associated with mutations that lead to an absence of dystrophin protein in muscle (2, 3). A clini- cally milder disease, Becket muscular dystrophy (BMD), accounts for ~20% of cases and often involves deletions within the rodlike central domain of dystrophin (4). Mus- cle dystrophin levels are often nearly normal in BMD, which can preclude diagnosis by immunohistochemical analysis ofdystrophin (5).

Dystrophin is a large intracellular protein containing several defined sequence motis (6). An NH2-terrninal cx-acti- nin-like domain binds to F-actin (7), and is followed by a large rod domain that shares sequence homology with the structural repeats in spectrin. The C O O H terminus is unique to dystrophin and related proteins, and this region

~ Abbreviations used in this paper: oe-BGT, alpha bungaro-toxin; AChR, ace- tylcholine receptor; BMD, Becket muscular dystrophy; DMD, Duchenne muscular dystrophy; GST, glutathione-S-transferase; nNOS, neural-type nitric oxide synthase; NO, nitric oxide; PH, pleckstrim homology.

direcdy binds to a glycoprotein complex in skeletal muscle (8-10). The structural dystrophin-associated complex in- cludes intracellular proteins, syntrophins (11), as well as in- tegral membranes proteins, the dystroglycans (12) and sarcoglycans; the absence of dystrophin in DMD causes a disruption of this complex (13). These interactions suggest a structural role for dystrophin, physically linking the extracellular matrix to the muscle cytoskeleton (14). In support of this model, genetic mutations in components of the sarcoglycan complex can cause autosomal recessive muscular dystrophy (15-18).

Restoration of a functional dystrophin molecule to muscle represents a primary goal for therapy. To better under- stand mechanisms for assembly of the dystrophin complex and to identify potential constructs for gene therapy, frag- ments of dystrophin have been targeted to skeletal muscle of transgenic mdx mice, which lack endogenous dystrophin. Replacement with either a full-length dystrophin, a COOH-terminal construct encoding 71 kD of dystrophin, Dp71, or a dystrophin minigene, lacking a large portion of the central spectrinlike repeats, restores the structural dystrophin complex to muscle. P..eplacement with full-length dystrophin corrects muscular dystrophy in mdx mice (19). Despite apparent restoration of the dystrophin complex, mdx mice targeted with Dp71, still display severe muscular

609 j. Exp. Med. �9 The Rockefeller University Press �9 0022-1007/96/08/609/10 $2.00 Volume 184 August 1996 609-618

dystrophy (8, 9), whereas those containing the minigene, have a very. mild disease phenotype (20, 21). These results indicate that all components o f the dystrophin membrane cytoskeleton are needed to completely prevent symptoms o f muscular dystrophy.

In addition to their cytoskeletal role, dystrophin and associated proteins have been implicated in specific signaling functions o f the junctional and extrajunctional sarcolemma. The dystrophin-related protein, utrophin, is concentrated at neuromuscular endplates and is implicated in acetylcho- line receptor (AChl<) clustering, ot-dystroglycan binds with high affinity to agrin and laminin suggesting that the dystrophin-associated complex may serve as a link between the extracellular matrix and intracellular events that help form AChR. clusters (22, 23). Signaling by the dystrophin complex may be mediated in part by nitric oxide (NO), a messenger molecule in muscle that can regulate myocyte development (24), AChR. function (25), and muscle contractility (26). N O is formed in skeletal muscle by the neuronal-type nitric oxide synthase (nNOS) that is enriched at the sarcolemma of fast twitch muscle fibers in rodents (26) and in both fast and slow twitch fibers in primates (27). R.ecent studies identify n N O S as a nonstructural compo- nent o f the dystrophin complex (28). Furthermore, n N O S is absent from skeletal muscle sarcolemma of mdx mice and in D M D (28). Biochemical studies in vitro demonstrate that the N H 2 terminus o f nNOS, which contains a PDZ protein motif, directly binds to a similar moti f in cel-syntrophin (29); furthermore, n N O S and od-syntrophin coimmunoprecipitate from muscle extracts. Direct binding of n N O S to dystrophin or other associated proteins has not yet been demonstrated.

D M D and mdx mice show primary dystrophin deficiency and secondary deficiencies o f sarcoglycans, dystroglycans, syntrophins, and nNOS. BMD in humans, due to abnormal dystrophin, generally retains dystrophin-associated proteins, though n N O S has not been evaluated. We now show that n N O S is properly restored to the plasma membrane in transgenic mdx mice expressing full-length human dystrophin but is selectively absent from skeletal muscle membranes in mdx mice expressing either Dp 71 (DMD phenotype) or a dystrophin minigene lacking many of the spectrinlike repeats (very mild BMD phenotype). Dyslocal- ization o f n N O S in the transgenic mouse nmdels is associated with disruption o f the normal nNOSAxl-syntrophin interaction. In human biopsies, we note that loss of sarcolemmal n N O S is commonly observed in BMD. In some patients, lacking as little as three exons in the spectrinlike domain of dystrophin, the absence of sarcolemmal nNOS represents the only identified immunohistochemical abnormality.

Materials and Methods Antibodies. The following primary antibodies were used:

nNOS polyclonal raised against homogenous nNOS protein purified from rat cerebellum (30), nNOS monoclonai (Transduction Labs, Lexington, KY), od-syntrophin polyclonal, syntrophin

monoclonal (31), dystrophin monoclonal (Signna Chemical Co., St. Louis, MO), [3-dystroglycan monoclonal, utrophin monoclonal, and a-sarcoglycan rnonoclonal (Novacastra Laboratories Ltd., Newcastle upon Tyne, UK).

lmmunofluorescence. Unfixed skeletal muscle samples were flash frozen in liquid nitrogen-cooled isopentane, sectioned on a cryostat (10 Ixm), and melted directly onto glass slides. Sections were then postfixed in 2% paraformaldehyde in PBS or cold ace- tone. Tissues were "blocked" in PBS containing 1% normal goat sermn. Primary antibodies were diluted in blocking reagent and were applied to sections overnight at 4~ For indirect immunofluorescence, secondary goat anti-rabbit FITC-, or donkey anti- mouse Cy-3-conjugated antibodies were used according to the manufacturer's specifications (The Jackson Laboratory, Bar Har- bor, ME; 1:200)_ Cy-3-conjugated 0~-BGT (kindly provided by Peter Sargent, University of California, San Francisco) was diluted together with the secondary antibody t-or double labeling motor endplates.

Tissue Extraction1 and Western Blot Analysis. Mouse quadriceps skeletal nmscle was homogenized in 1(1 volumes (wt/vol) of buffer A (25 mM Tris-HC1, pH 7.4, 100 mM NaCl, 1 mM EDTA, l mM EGTA, and 1 mM PMSF), and heavy, microsomes were prepared by a standard protocol with minor modifications. Nuclei were pelleted by centrifugation at 1,000 ,~. The supernatant was then centrifuged at 20,000 ,~, yielding supernatant S~. The resulting heavy microsomal pellet was resuspended in buffer A containing 500 mM NaC1, incubated for 30 rain at 4~ with agitation, and centrifuged at 15,000 ,~, yielding supernatant S,. This resulting pellet was resuspended in buffer A containing 5~0 mM NaC1 plus 0.5% Triton X-100, incubated for 30 mm at 4o( ", with agitation, and centrifuged at 15,000 ,~, yielding supernatant $3 and a final pellet P.

Tissue extracts were resolved by SDS-PAGE (7.5% acryla- mide) and proteins were transferred to polyvinylidene fluoride membranes (hnmobilon-P; Millipore Corp., Bedford, MA). Membranes were incubated overnight with primary antisera diluted in Tris-HC1 buffered saline containing 1% BSA, hnnmnoreactive bands were visualized by the enhanced chemiluminescence sys- tem according to the manufacturer's specifications (Amersham Corp., Arlington Heights, ILl.

Immunoprecipitation. Polyctonal antibodies (1 ~g) to od-syntrophin or noninmmne serum were added to 0.5-ml aliquots of solubilized skeletal muscle membranes from wild-type mouse (1 mg/ml) or total solubilized nmscle extract from mdx mouse (2 mg/ml), and samples were incubated on ice fbr 1 h. Protein A-Sepharose (50 p,1) was used to precipitate antibodies. Protein A pellets were washed three times with buffer containing I{t0 mM NaCI and 1% Triton X-100. Immunoprecipitated proteins were denatured with loading buffer and resolved by SDS-PAGE.

Fusion P~vtein Aflb~ity Chromatq~raphl,. A thsion protein of glutathione-S-transferase (GST) fused to the first 299 amino acids of nNOS was expressed in Escheridzia coil and purified on glutathione Sepharose beads as described (28). Solubilized skeletal muscle membranes were incubated with control ((;ST) or GST- nNOS (1-299) beads. Samples were loaded into disposable col- unms washed with 5(1 volumes of buffer containing 0.5% Triton X-I(10 plus 3(}(I mM NaC1, and proteins were eluted with 150 p,1 of SDS-PAGE loading buffer.

Characterization qfHumal~ Tissues. All human muscle biopsies were obtained for diagnostic purposes (dystrophin analysis), and were flash frozen in isopentane cooled in liquid nitrogen. Patients were evaluated for dystrophin expression by immunofluorescence and Western blotting as described (32, 33). Mutation detection in

610 Sarcolemmal Nitric Oxide Synthase in Becker Muscular Dystrophy

BMD patients was done by multiplex PCR, as previously described. Mutation detection in or (adhalin) was done by R T - P C R and single strand conformation polymorphism, with aberrant conformers sequenced as previously described (34); one patient was homozygous for an Arg77Cys mutation whereas another was a compound heterozygote, Leu31Pro and Arg284Cys. All biopsies used in this study were deemed to be of excellent preservation based on hematoxylin and eosin staining of cryosections.

Results

Previous studies suggest that association o f n N O S with the dystrophin complex is mediated by direct binding o f

the NH2-terminus o f n N O S to the P D Z domain o f c~l- syntrophin (29). However , during skeletal muscle development, we found a dissociation between n N O S and oel-syn- t rophin localization in muscle. Throughout postnatal rat development , o t l -syntrophin was present at extrajunctional sarcolemma and was particularly enriched at neuromuscular endplates (Fig. 1). By contrast, at postnatal day 3 (P3) and P7, n N O S was observed only at extrajunctional sarcolemma. Enr ichment o f nNOS at neuromuscular endplates did not become apparent until P12, coincident with accu- mulat ion o f dystrophin at endplates. Ut rophin was en- r iched at endplates in all stages evaluated.

W e compared n N O S and ot l -syntrophin expression in

Figure 1. Localization of nNOS and other dystrophin-associated proteins during postnatal development. Adjacent sections of postnatal rat quadriceps muscle were stained for dystrophin, nNOS, oe-BGT, and utrophin, and nearby sections were stained for otl-syntrophin and cx-BGT. Dystrophin and uNOS stained extrajunctional sarcolemma at P3 and P7 and both became concentrated at neuromuscular endplates at P12 and P60. otl-syntrophin was present at extrajunctional sarcolemma and was enriched at neuromuscular endplates at all ages evaluated. Utrophin staining was restricted to neuromuscular endplates.

611 Chao et al,

Figure 2. Localization of nNOS and ~xl-syntrophin in transgenic mdx mice. Cryosections from mouse quadriceps were double labeled for either nNOS or ~xl-syntrophin and cr Immunofluorescent staining showed that nNOS in wild-type mouse was expressed at extrajunctional sareolemma of a subset of fibers and was enriched at all neuromuscular endplates, nNOS was absent from junctional and extrajunctional sarcolemma in mdx mice. nNOS staining in mdx transgenic mice expressing full-length dystrophin (full-dys) or truncated dystrophin lacking the COOH-termina1330 nucleotides (A331)) resembled that of wild-type mice. mdx mice expressing dystrophin-lacking exons 17--48 (AEXON 17-48 mini-dys) or the COOH-terrmna171 kD of dystrophin Dp7l lacked nNOS staining at sarcolemma, similar to nontransgenic mdx mouse, otl-syntrophin occurred at extrajunctional sarcolemma a~ad was concentrated at neuromuscular endplates in wild-type mice and was restricted to the endplates in mdx mouse, cd-syntrophin expression was restored to sarcolemma in the four transgenic mdx mouse lines expressing different portions of the dystrophin gene.

skeletal muscle o f wi ld- type , mdx, and various transgenic mdx mice (Fig. 2) that express mu tan t forms of dystrophin (8, 9, 19-21, 35). As previously reported, o t l - syn t roph in was absent f rom extrajunct ional sarcolemma o f mdx mouse , bu t remained at neuromuscu la r endplates (31, 36). n N O S was absent f rom bo th junc t iona l and extrajunct ional sarcolemma o f mdx mouse. Four lines of mdx transgenic mice

were evaluated. These previously described lines express either ful l - length dystrophin (line 8 6 2 C A A [201), or various t runcated dystrophins, lacking either the 330 nucleot ides of exons 7 1 - 7 4 near the C O O H - t e r m i n u s (A330 [35]), lacking exons 17-48 of the spectrinlike mo t i f (AE17-48; C V B A line 12142 [20]), or lacking all bu t the C O O H - t e r - minal 71 kD (Dp71; M C A - 1 [8, 9]). As previously re-


Figure 3. Subcellular distribution of nNOS in transgenic mdx mouse. Mouse quadriceps skeletal muscle homogenates were sequentially extracted with buffers containing 100 mM NaC1 (S1), 500 mM NaCI ($2), and 0.5% Triton X-100 ($3), leaving an insoluble cytoskeletal pellet (P). (A) Western blotting indicated that nNOS was enriched in membrane- associated and pellet fractions in wild-type mouse (lanes 1) and transgenic mdx mice expressing full-length dystrophin (lanes 4). In mdx mouse (lanes 2) and Dp71 transgenic mdx mouse (lanes 3), nNOS was fully extracted by 500 mM NaCI and was absent from membrane-associated and cytoskeletal fractions. (B and C) A similar fractionation was performed on muscle homogenates from wild-type nmuse (lanes 1), mdx mouse expressing dystrophin-lacking exons 17-48 (lanes 2), or mdx mouse expressing fill-length dystrophin (lanes 3). (B) nNOS was absent from membrane- associated ($3) and cytoskeletal pellet (P) in mdx mouse expressing the truncated dystrophin. (C) R.eprobing the blot shows that od-syntrophin had a generally similar fractionation in nmscle from all three mice lines.

p o r t e d , o L l - s y n t r o p h i n e x p r e s s i o n was r e s t o r e d to j u n c -

t iona l a n d e x t r a j u n c t i o n a l s a r c o l e m m a in e a c h o f t h e f o u r

t r ansgen ics . B y con t ra s t , n N O S was r e s t o r e d to t h e s a r c o -

l e m m a o n l y b y f u l l - l e n g t h d y s t r o p h i n a n d t h e A 3 3 0 m u -

tant .

Figure 4. Selective interaction o fnNOS and od-syntrophin. Crude solubilized membranes from mouse quadriceps were titrated with NaOH to pH 11, to dissociate the dystrophin complex, and were neutralized to pH 7.4 with 1 M Tris-HC1. Native and dissociated (&~soc) preparations were incu-

bated with agarose beads linked to either GST or GST fused to the first 299 anfino acids ofnNOS (G-NOS). After extensive washing, beads were eluted with loading buffer and proteins were resolved by SDS/PAGE. (A) West- em blotting showed that txl-syntrophin was selectively retained by G-NOS beads in both native and dissociated preparations. Reprobing the same blot with (B) dystrophin or (C) ot-sarcoglycan revealed that G-NOS beads retained these proteins from native protein preparations. However, after dissociation of the complex, neither dystrophin nor ot-sarcoglycan bound to G-NOS. The 55-kD band observed in input lanes from r blot appears to be mouse IgG and was reactive with the secondary antibody used for Western blotting. (D) Coimmunoprecipitation ofnNOS with otl-syntrophin from wild-type and mdx mouse skeletal muscle. Total solubilized extract from mdx (lanes 1 and 2) or solubilized membranes from wild-type (lanes 3 and 4) mouse quadriceps were immunoprecipitated with an antibody to od-syntrophin (lanes 1 and 3) or nonimmune serum (lanes 2 and 4). Western blotting indicates that nNOS was specifically coimmu- noprecipitated with oel-syntrophin from mdx and wild-type extracts.

613 Chao et al.

Table 1. Sarcolemmal Expression of nNOS, Dystrophin, and Syntrophin in BMD

Imunofluorescence at sarcolenlnla Exons deleted

Diagnosis (mutation) Age at biopsy nNOS Dystrophin Syntrophin

Normal 53 + + + + + + + + + + + +

Normal 41 + + + + + + + + + + + +

Mild BMD 45-47 30 0 + + + + + + + +

Mild BMD 52 29 + + + + + + + + + + +

Mild BMD 45-48 8 + + + + + + + + +

Int. BMD 3-6 13 + + + + + + + +

Int. BMD 10-42 24 0 + + + + + + + +

Int. BMD 13-41 12 + + + + + + + + +

Int. BMD 45 23 0 + + +

Sev. BMD 8 10 0 + + + +

Sev. BMD 3-7 7 0 + + + +

Sev. BMD 45-47 t0 0 + + + + + + +

Sev. BMD 51-52 9 (I + + + +

a-sarcoglycanopathy (L31P/R284C) l l + + + + + + + + + + + +

{x-sarcoglycanopathy (P,77C/P,.77C) 8 + + + + + + + + + +

Human muscle biopsies were labeled by immunofluorescence. Sarcolemmal labeling was blindly evaluated by three observers from {) to + + + +. Variation between observers never varied by more than one +, and for those cases, the majority score is reported, bit., intermediate; Sev., severe.

Biochemical studies confirmed that n N O S did not associate with sarcolemma in mdx mice or transgenic mdx mice expressing either Dp71 or ~E17-48. In wi ld- type and mdx transgenic mice expressing full-length dystrophin, n N O S was enriched in membrane-associated and cytoskeletal fractions, whereas in mdx, Dp71, and ZXE17-48 lines, n N O S was present only in soluble fractions of muscle. As previously reported, ed-synt rophin occurred in sarcolemmal fractions of all four lines o f transgenic mdx mice evaluated (Fig. 3, and data not shown).

Whereas these studies are consistent with the model that association of n N O S with ed-synt rophin in vivo requires a full-length dystrophin, direct binding o f n N O S to the rodlike domain of dystrophin or another dystrophin-associated protein could also explain the data. Previous studies demonstrate that the NH2-terminal domain o f n N O S is necessary and sufficient for interaction with the dystrophin complex (28). W e therefore evaluated interact ion o f dys t rophin- associated proteins with a purified fusion protein containing the first 299 amino acids o f n N O S . As previously demonstrated, a Sepharose column linked to this fusion protein selectively retained several components of the dystrophin complex from crude skeletal muscle extracts. To determine which components directly interact with n N O S in vitro, we dissociated the dystrophin complex by briefly adjusting the pH of muscle extracts to 11 and then repeating the

binding assays immediately after neutralizing the extracts. Previous studies (37) have demonstrated that this procedure reversibly dissociates dystrophin from associated proteins. After dissociation o f the complex, ed-synt rophin contin- ued to interact with the n N O S column but dystrophin, tx-sarcoglycan, and [3-dystroglycan were not retained (Fig. 4, A - C and data not shown).

To further verify that n N O S directly interacts with syntrophin in vivo, we conducted inmmnoprecipi ta t ion ex- periments in mdx mouse (Fig. 4/7)). A polyclonal antibody to ed-synt rophin specifically coimnmnoprecipi ta ted a small amount of n N O S from extracts of mdx skeletal nmscle. n N O S was more efficiently coinmmnoprecipi tated with e~ 1- syntrophin from solubilized membranes of wild-type mouse as previously shown (29).

W e next asked whether mutations in the NH2-terminal or rodlike domains of dystrophin that cause B M D in humans were associated with altered localization of nNOS. W e immunohistochemical ly evaluated n N O S and txl-syntrophin expression in 12 B M D patients with molecularly defined deletions in the dystrophin gene. Immunohis- tochemical expression o f nNOS, dystrophin, and syntrophin were assessed blindly. Loss ofsarcolemmal n N O S , but not cd-synt rophin expression was associated with Becker phenotype (Table 1). Some o f the patients, with mild to intermediate disease, showed reduced but detectable n N O S


Figure 5. nNOS is absent from skeletal muscle sarcolemma in certain patients with BMD. Skeletal muscle cryosections from human biopsies were im- munostained with monoclonal antibodies to dystrophin, syntrophin, ot-sarcoglycan, or polyclonal antibody nNOS. All four antibodies showed sarcolemmal staining in normal patients and essentially no sarcolemmal labeling in patients with DMD. In two patients with BMD, due to loss of exons 45-47 or 10-42 of dystrophin, immunofluorescent labeling for dystrophin, syntrophin, and ot-sarcoglycan was detected at the membrane. By contrast, nNOS sarcolemmal staining was undetectable in these two BMD patients, nNOS labeling was present in a patient with ot-sarcoglycan deficiency.

staining ofsarcolemma. In several patients, loss ofsarco lem- mal n N O S occurred despite apparently normal assembly o f other components o f the dystrophin-associated glycoprotein complex (Fig. 5). By contrast, we found that n N O S expression was intact in two patients with primary et-sarcoglycan deficiency. This is consistent with the normal sta- tus o f dystrophin and syntrophins in this disorder.

Discuss ion

A principal finding o f this work is that assembly o f n N O S into the dystrophin complex is dependent upon the normal structure o f the rodlike domains ofdyst rophin . Pre- vious analyses o f specific protein contacts involved in main- tenance o f the dystrophin complex have focused on protein

615 Chao et al.

interactions at the N H 2- and C O O H - t e m n i n a l domains o f dystrophin. These studies identify a functional F-actin binding site near the N H a - t e r m i n u s (7) and b ind ing sites for 13-dystroglycan and syntrophins in the C O O H - t e r m i - nal domain o f dystrophin (10). Understanding the mecha- nism for n N O S association with the dystrophin complex is important because n N O S is uniquely absent from sarcolemma in certain animal models o f muscular dystrophy and in certain patients with BMD. Absence o f sarcolemmal n N O S in mdx mouse expressing a dystrophin minigene indicates a role for the rodlike domain o f dystrophin for binding o f n N O S . Studies o f n N O S expression in B M D patients demonstrate that distinct deletions in the NH2- te r - minal or central domain o f dystrophin disrupt recrui tment o f n N O S to the sarcolemma. These results indicate that a

unique nNOS interaction domain may not be present in dystrophin, but that proper conformation is required for assembly o fnNOS into the dystrophin complex.

Previous studies suggest that direct interaction of nNOS with 0d-syntrophin accounts for association of nNOS with the dystrophin complex (29). Three syntrophin genes have been identified and each contains two pleckstrin homology (PH) domains. The first PH domain is split by a PDZ motif, and the second PH domain is followed by a C O O H - terminal region unique to the syntrophins (38). Interaction of nNOS with ~xl-syntrophin is mediated by direct association of PDZ protein-binding interfaces near the NH2-terminus of nNOS and cxl-syntrophin. Studies here are consistent with this model and demonstrate that &l-syntrophin, but not dystrophin, [3-dystroglycan or cl-sarcoglycan binds to the PDZ-containing domain of nNOS. NOS isoforms lacking a PDZ motif do not associate with the dystrophin complex, further suggesting that the PDZ domain of nNOS represents the relevant domain for interaction (29).

The precise binding site(s) for syntrophins within the dystrophin complex is uncertain. In vitro studies show that the COOH-terminal region of syntrophin directly interacts with a splice-prone COOH-terminal domain ofdystrophin (39-41). However, syntrophins are present in dystrophin complexes of the A330 transgenic mouse that lacks the identified syntrophin interaction domain, suggesting addi- tional bindings sites for syntrophins (35). It is not clear which domain of syntrophins might interact with these ad- ditional sarcolemmal binding sites. PH domains are known to interact with specific membrane proteins and phospho- lipids (42), and these regions of syntrophins represent can- didate interaction domains. Because the first PH domain of syntrophin is split by the PDZ domain (38), it is possible that sinmltaneous occupation of the PH1 and PDZ sites is sterically prohibited. Therefore, membrane association of syntrophin mediated by its PH domains could preclude binding of nNOS to the PDZ domain. This may explain the observed sarcolemmal expression of syntrophin and absence o fnNOS in certain disease states.

nNOS appears not to interact with utrophin-containing complexes. During early postnatal muscle development, nNOS is not concentrated with complexes of cxl-syntrophin and utrophin at neuromuscular endplates. Similarly, utrophin complexes at neuronmscular endplates of mdx mouse lack nNOS. Taken together with biochemical studies showing direct interaction of nNOS with Rl-syntrophin in vitro, we propose that sarcolemmal localization of nNOS requires both syntrophin and dystrophin. It is alter- natively possible that nNOS primarily binds directly to dystrophin in vivo. We have, however, been unable to detect direct interaction between nNOS and dystrophin in vitro and would disfavor this model. Future studies o fnNOS expression in mice lacking syntrophin isoforms may be necessary to definitively clarify this issue.

Loss of sarcolemmal nNOS does not appear to be a ge- neric consequence of muscle disease, nNOS expression oc- curs normally at the sarcolemma in a variety of inflamma- tory, neuropathic, and idiopathic muscle disorders (Chao, D.S., and D.S. Bredt, unpublished observations), and in dy mutant mice that have muscular dystrophy associated with loss of extracellular M-laminin (merosin; 28). Three patients with similar deletions of exons 45-47 of dystrophin all showed loss of sarcolemmal nNOS but exhibited a wide spectrum of clinical variability. This domain of dystrophin is apparently critical for assembly ofnNOS, and the clinical differences could be caused by environmental and/or epi- genetic factors, nNOS was expressed normally in two patients with autosomal recessive muscular dystrophy due to mutations in oL-sarcoglycan (adhalin). Abnormality of nNOS expression, therefore, appears specific for dystrophin-related disease, and immunohistochernical analysis for nNOS may provide a supplemental diagnostic test.

Abnormal expression of nNOS may play a role in the pathophysiology of BMD. Decreased expression of nNOS alone is not sufficient to produce muscular dystrophy, as we have not detected muscle pathology in nNOS a/a mice that have a targeted disruption o fn N O S (Chao, D.S., and D.S. Bredt, unpublished observations). However, endogenous NO does play a role in regulation of skeletal muscle development and contractility (24, 26). Disruption of these NO signaling pathways may contribute to abnormal nmscle function and incomplete myofiber regeneration seen in muscular dystrophy. In mdx mice expressing Dp71 or dystrophin minigene, nNOS is the only known dystrophin- associated protein absent from the sarcolemnaa. Transgenic mdx mice expressing the dystrophin minigene have an ex- tremely mild muscular dystrophy characterized by only a modest increase in central nuclei and serum pyruvate kinase activity (20, 21). A human patient with an identical mutation had a mild dystrophy and was ambulatory with the aid of a stick at age 61 (43). In contrast to the mild clinical phenotype, this minigene mutation in a 25-year-old patient was associated with severe histopathological nmscle fiber atrophy, extensive replacement by fat and fibrous connec- tive tissue, and few surviving fibers of normal diameter (43).

These findings may be relevant in designing therapies for DMD. We find that nNOS does not associate with utrophin-containing complexes, so that strategies for upregula- tion ofutrophin in DMD would not be expected to restore sarcolemmal nNOS. Because of the large size of the dystrophin protein, vectors for gene therapy may need to encode truncation mutants. Expression ofa minigene lacking exons 17-48 of dystrophin fails to recruit nNOS to sarcolemma of mdx mouse and is associated with a very mild Becket phenotype. Complete rescue of muscle function may re- quire replacement with dystrophin constructs that properly recruit nNOS to skeletal muscle membranes.

This work was supported by grants from the Muscular Dystrophy Association, the National Institutes of Health (NIH) 1KOINS34822, the Council for Tobacco Research, the Smokeless Tobacco Research Coun-

616 Sarcolermnal Nitric Oxide Synthase in Becker Muscular Dystrophy

cil, and the Lucille P. Markey Charitable Trust (to D.S. Bredt); from the Parent Project (to E.P. Hoffman); and from the NIH and the Muscular Dystrophy Association (to S.C. Froehner),

Address correspondence to Dr. David S. Bredt, Department of Physiology and Program in Biomedical Sci- ences, University of California at San Francisco School of Medicine, 513 Parnassus Avenue, San Francisco, CA 94143-0444.

Received for publication 16 April 1996 and in revised form 8 May 1996.

References

1. Engel, A.G., and B.Q. Banker. 1994. Myology: Basic and Clinical. McGraw-Hill Inc., New York. 215a pp.

2. Hoffman, E.P. 1993. Genotype/Phenotype Correlations in Duchenne/Becker Muscular Dystrophy. Mol. Cell Biol. Hum. Dis. 3:12-36.

3. Hoffman, E.P., K.H. Fischbeck, lL.H. Brown, M. Johnson, R. Medori, J.D. Loike, J.B. Harris, R. Waterston, M. Brooke, L. Specht et al. 1988. Characterization ofdystrophin in muscle-biopsy specimens from patients with Duchenne's or Becker's muscular dystrophy. N. Enql. J. Med. 318:1363- 1368.

4. Beggs, A.H., E.P. Hoffman, J.R. Snyder, K. Arahata, L. Specht, F. Shapiro, C. Angelini, H. Sugita, and L.M. Kunkel. 1991. Exploring the molecular basis for variability among patients with Becker muscular dystrophy: dystrophin gene and protein studies. Am. J. Hum. Genet, 49:54-67.

5. Arahata, K., E.P. Hoffman, L.M. Kunkel, S. Ishiura, T. Tsu- kahara, T. Ishihara, N. Sunohara, I. Nonaka, E. Ozawa, and H. Sugita. 1989. Dystrophin diagnosis: comparison ofdystrophin abnormalities by immunofluorescence and immunoblot analyses. Proc. Natl. Acad. Sci. USA. 86:7154-7158.

6. Koenig, M., A.P. Monaco, and L.M. Kunkel. 1988. The complete sequence of dystrophin predicts a rod-shaped cytoskeletal protein. Cell. 53:219-26.

7. Way, M., B. Pope, R.A. Cross, J. Kendrick-Jones, and A.G. Weeds. 1992. Expression of the NH2-terminal domain of dystrophin in Escherichia coli and demonstration of binding to F-actin. FEBS (Fed. Eur. Biochem. Soc.) Lett. 301:243-245.

8. Cox, G.A., Y. Sunada, K.P. Campbell, andJ.S. Chamberlain. 1994. Dp71 can restore the dystrophin-associated glycopro- rein complex in muscle but fail to prevent dystrophy. Nature Genetics. 8:333-339.

9. Greenberg, D.S., Y. Sunada, K.P. Campbell, D. Yaffe, and U. Nudel. 1994. Exogenous Dp71 restores the levels ofdystrophin associated proteins but does not alleviate muscle damage in mdx mice. Nature Genetics. 8:340-344.

10. Suzuki, A., M. Yoshida, K. Hayashi, Y. Mizuno, Y. Hagi- wara, and E. Ozawa. 1994. Molecular organization at the glycoprotein-complex-binding site of dystrophin. Three dystrophin-associated proteins bind directly to the carboxy-terminal portion ofdystrophin. Eur. J. Biochem. 220:283-292.

11. Adams, M.E., M.H. Butler, T.M. Dwyer, M.F. Peters, A.A. Murnane, and S.C. Froehner. 1993. Two forms of mouse syntrophin, a 58 kd dystrophin-associated protein, differ in primary structure and tissue distribution. Neuron. 11:531-540.

12. Ibraghimov-Beskrovnaya, O., J.M. Ervasti, C.J. Leveille, C.A. Slaughter, S.W. Semett, and K.P. Campbell. 1992. Pri- mary structure of dystrophin-associated glycoproteins linking dystrophin to the extracellular matrix. Nature (Lond.). 355: 696-702.

13. Ervasti, J.M., K. Ohlendieck, S.D. Kahl, M.G. Gaver, and

617 Chao et al.

K.P. Campbell. 1990. Deficiency of a glycoprotein compo- nent of the dystrophin complex in dystrophic muscle. Nature (Lond.). 345:315-319.

14. Ervasti, J.M., and K.P. Campbell. 1993. Dystrophin and the membrane skeleton. Curt. Opin. Cell. Biol. 5:82-87.

15. Bonnemann, C.G., R. Modi, S. Noguchi, Y. Mizuno, M. Yoshida, E. Gussoni, E.M. McNally, D.J. Duggan, C. Ange- lini, E.P. Hoffman et al. 1995. Beta-sarcoglycan (A3b) mutations cause autosomal recessive muscular dystrophy with loss of the sarcoglycan complex. Nature Genetics 11:266-273.

16. Lim, L.E., F. Duclos, O. Broux, N. Bourg, Y. Sunada, V. A1- lamand, J. Meyer, I. Richard, C. Moomaw, C. Slaughter et al. t995. Beta-sarcoglycan: characterization and role in limb- girdle muscular dystrophy linked to 4q12. Nature Genetics. 11: 257-265.

17. Noguchi, S., E.M. McNally, K. Ben Othmane, Y. Hagiwara, Y. Mizuno, M. Yoshida, H. Yamamoto, C.G. Bonnemann, E. Gussoni, P.H. Denton et al. 1995. Mutations in the dystrophin-associated protein gamma-sarcoglycan in chromosome 13 muscular dystrophy [see comments]. Science (Wash. DC). 270:819-822.

18. Roberds, S.L., F. Leturcq, V. Allamand, F. Piccolo, M. Jean- pierre, lL.D. Anderson~ L.E. Lira, J.C. Lee, F.M. Tome, N.B. Romero, et al. 1994. Missense mutations in the adhalin gene linked to autosomal recessive muscular dystrophy. Cell. 78:625-633.

19. Cox, G.A., N.M. Cole, K. Matsumura, S.F. Phelps, S.D. Hauschka, K.P. Campbell, J.A. Faulkner, and J.S. Chamber- lain. 1993. Overexpression of dystrophin in transgenic mdx mice eliminates dystrophic symptoms without toxicity [see comments]. Nature (Lond.). 364:725-729.

20. Phelps, S.F., M.A. Hauser, N.M. Cole, J.A. Rafael, R,T. Hinkle, J.A. Faulkner, and J.S. Chamberlain. 1995. Expres- sion of full-length and truncated dystrophin mini-genes in transgenic mdx mice. Hum. Mol. Genet. 4:1251-1258.

21. Wells, D.J., K.E. Wells, E.A. Asante, G. Turner, Y. Sunada, K.P. Campbell, F.S. Walsh, and G. Dickson. 1995. Expres- sion of human full-length and minidystrophin in transgenic mdx mice: implications for gene therapy of Duchenne muscular dystrophy. Hum. Mol. Genet. 4:1245-1250.

22, Campanelli, J.T., S.L. Roberds, K.P. Campbell, and R.H. Scheller. 1994. A role for dystrophin-associated giycoproteins and utrophin in agrin-induced AChR clustering. Cell. 77: 663-674.

23. Gee, S.H., F. Montanaro, M.H. Lindenbaum, and S. Car- bonetto. 1994. Dystroglycan-alpha, a dystrophin-associated glycoprotein, is a functional agrin receptor. Cell. 77:675-686.

24. Lee, K.H., M.Y. Baek, K.Y. Moon, W.K. Song, C.H. Chung, D.B. Ha, and M.S. Kang. 1994. Nitric oxide as a messenger molecule for myoblast fusion. J. Biol. Chem. 269: 14371-14374.

25. Wang, T., Z. Xie, and B. Lu. 1995. Nitric oxide mediates activity-dependent synaptic suppression at developing neuromuscular synapses. Nature (Lond.). 374:262-266.

26. Kobzik, L., M.B. Reid, D.S. Bredt, and J.S. Stamler. 1994. Nitric oxide in skeletal muscle. Nature (Lond.). 372:546-548.

27. Grozkanovic, Z., G. Nakos, G. Dahramann, B. Mayer, and P,. Gossrau. 1995. Species-independent expression of nitric oxide synthase in the sarcolemma region of visceral and somatic striated muscle fibers. Cell Tissue Res. 281:493-499.

28. Brenman, J.E., D.S. Chao, H. Xia, K. Aldape, and D.S. Bredt. 1995. Nitric oxide synthase complexed with dystrophin and absent from skeletal muscle sarcolemma in Duch- enne muscular dystrophy. Cell. 82:743-752.

29. Brenman, J.E., D.S. Chao, S.H. Gee, A.W. McGee, S.E. Craven, D.R. Santillano, F. Huang, H. Xia, M.F. Peters, S.C. Froehner, and D.S. Bredt. 1996. Interaction of nitric oxide synthase with the synaptic density protein PSD-95 and or-1 syntrophin mediated by PDZ motifs. Cell. 84:757-767.

30. Bredt, D.S., P.M. Hwang, and S.H. Snyder. 1990. Localiza- tion of nitric oxide synthase indicating a neural role for nitric oxide. Nature (Lond.). 347:768-770.

31. Peters, M.F., N.R. Kramarcy, P,.. Sealock, and S.C. Froeh- her. 1994. Beta 2-Syntrophin: localization at the neuromuscular junction in skeletal muscle. Neuroreport. 5:1577-1580.

32. Hoffman, E.P., L.M. Kunkel, C. Angelini, A. Clarke, M. Johnson, andJ.B. Harris. 1989. Improved diagnosis of Becker muscular dystrophy by dystrophin testing. Neurology. 39: 1011-1017.

33. Hoffman, E.P., J.E. Morgan, S.C. Watkins, and T.A. Par- tridge. 1990. Somatic reversion/suppression of the mouse mdx phenotype in vivo. J. Neurol. &i. 99:9-25.

34. Duggan, D.J., M. Fanin, E. Pegoraro, C. Angelini, and E.P. Hoffman. 1996. Adhalinopathies: complete biochemical deficiency patients are 5% of childhood-onset dystrophin-normal muscular dystrophy and most partial deficiency patients do

not have gene mutations. J. Neurol. Sci. In press. 35. Rafael, J.A., Y. Sunada, N.M. Cole, K.P. Campbell, J.A.

Faulkner, and J.S. Chamberlain. 1994. Prevention of dystrophic pathology in mdx mice by a truncated dystrophin iso- form. Hum. Mol. Genet. 3:/725-1733.

36. Yang, B., O. IbraghimowBeskrovnaya, C.R.. Moomaw, C.A. Slaughter, and K.P. Campbell. 1994. Heterogeneity of the 59-kDa dystrophin-associated protein revealed by eDNA clon- ing and expression.J. Biol. Chenl. 269:6040-6044.

37. Ervasti, J.M., S.D. Kahl, and K.P. Campbell. 1991. Purifica- tion of dystrophin from skeletal muscle. J. Biol. Chem. 266: 9161-9165.

38. Adams, M.E., T.M. Dwyer, L.L. Dowler, R.A. White, and S.C. Froehner. 1995. Mouse ~xl- and [~2-syntrophin gene structure, chromosome localization, and homology with a discs large domain.J. Biol. Chem. 270:25859-25865.

39. Ahn, A.H., and L.M. Kunkel. 1995. Syntrophin binds to an al- ternatively spliced exon of dystrophin. J. Cell Biol. 128:363-371.

40. Suzuki, A., M. Yoshida, and E. Ozawa. 1995. Mammalian alpha I- and beta 1-syntrophin bind to the alternative splice- prone region of the dystrophin COOH terminus. J. Cell. Biol. 128:373-381.

41. Yang, B., D. Jung, J.A. Rafael, J.S. Chamberlain, and K.P. Campbell. 1995. Identification of alpha-syntrophin binding to syntrophin triplet, dystrophin, and utrophin. J. Biol. Chem. 270:4975-4978.

42. Inglese, J., W.J. Koch, K. Touhara, and P,.J. Lefkowitz. 1995. G beta gamma interactions with PH domains and Ras- MAPK signaling pathways. Trends Biochem. Sci. 20:151-156.

43. England, S.B., L.V. Nicholson, M.A. Johnson, S.M. Forrest, D.R. Love, E.E. Zubrzycka-Gaarn, D.E. Bulman, J.B. Har- ris, and K.E. Davies. 1990. Very mild muscular dystrophy associated with the deletion of 46% of dystrophin. Nature (Lond.). 343:18(}-I 82.


selective loss of sarcolemmal nitric oxide synthase in becker muscular dystrophy

Documents